深度学习在基于曲面体层片的成釉细胞瘤及牙源性角化囊肿鉴别诊断中的应用  

Application of Deep Learning in Differential Diagnosis of Ameloblastoma and Odontogenic Keratocyst Based on Panoramic Radiographs

作　　者：李敏 慕创创[1,2] 张建运 李刚 LI Min;MU Chuangchuang;ZHANG Jianyun;LI Gang(Department of Oral and Maxillofacial Radiology,Peking University Hospital of Stomatology,Peking University School of Stomatology,Beijing 100081,China;Beijing Key Laboratory of Digital Stomatology,National Engineering Research Center of Oral Biomaterials and Digital Medical Devices,National Clinical Research Center for Oral Diseases,Beijing 100081,China;Department of Medical Pathology,Peking University Hospital of Stomatology,Peking University School of Stomatology,Beijing 100081,China)

机构地区：[1]北京大学口腔医学院口腔医院医学影像科,北京100081 [2]国家口腔疾病临床医学研究中心、口腔生物材料和数字诊疗装备国家工程研究中心、口腔数字医学北京市重点实验室,北京100081 [3]北京大学口腔医学院口腔医院医学病理科,北京100081

出　　处：《中国医学科学院学报》2023年第2期273-279,共7页Acta Academiae Medicinae Sinicae

基　　金：北京大学百度基金(2020037)。

摘　　要：目的通过应用不同卷积神经网络模型对成釉细胞瘤及牙源性角化囊肿进行鉴别诊断。方法回顾性收集1000张成釉细胞瘤和牙源性角化囊肿患者的数字曲面体层片,选用ResNet(18、50、101)、VGG(16、19)、EfficientNet(b1、b3、b5)深度学习模型,对训练集中的800张曲面体层片经五折交叉验证的方法训练后对测试集中的200张曲面体层片进行鉴别诊断。同时,7名口腔放射专业医生对这200张曲面体层片进行诊断,并将二者的诊断结果进行比较。结果卷积神经网络模型的诊断准确率为82.50%~87.50%,其中EfficientNet b1准确率最高,为87.50%,各卷积神经网络模型训练集和测试集本身之间比较,准确率差异无统计学意义(P_(训练集)=0.998,P_(测试集)=0.905)。7名口腔放射专业医生(2名高年资医生、5名低年资医生)平均诊断准确率为(70.30±5.48)%,且不同年资医生之间平均诊断准确率差异无统计学意义(P=0.883)。深度学习卷积神经网络模型的诊断准确率显著高于口腔放射专业医生的诊断准确率(P<0.001)。结论基于曲面体层片的深度学习卷积神经网络能够对成釉细胞瘤和牙源性角化囊肿做出较为准确的鉴别诊断。Objective To evaluate the accuracy of different convolutional neural networks(CNN),representative deep learning models,in the differential diagnosis of ameloblastoma and odontogenic keratocyst,and subsequently compare the diagnosis results between models and oral radiologists.Methods A total of 1000 digital panoramic radiographs were retrospectively collected from the patients with ameloblastoma(500 radiographs)or odontogenic keratocyst(500 radiographs)in the Department of Oral and Maxillofacial Radiology,Peking University School of Stomatology.Eight CNN including ResNet(18,50,101),VGG(16,19),and EfficientNet(b1,b3,b5)were selected to distinguish ameloblastoma from odontogenic keratocyst.Transfer learning was employed to train 800 panoramic radiographs in the training set through 5-fold cross validation,and 200 panoramic radiographs in the test set were used for differential diagnosis.Chi square test was performed for comparing the performance among different CNN.Furthermore,7 oral radiologists(including 2 seniors and 5 juniors)made a diagnosis on the 200 panoramic radiographs in the test set,and the diagnosis results were compared between CNN and oral radiologists.Results The eight neural network models showed the diagnostic accuracy ranging from 82.50%to 87.50%,of which EfficientNet b1 had the highest accuracy of 87.50%.There was no significant difference in the diagnostic accuracy among the CNN models(P=0.998,P=0.905).The average diagnostic accuracy of oral radiologists was(70.30±5.48)%,and there was no statistical difference in the accuracy between senior and junior oral radiologists(P=0.883).The diagnostic accuracy of CNN models was higher than that of oral radiologists(P<0.001).Conclusion Deep learning CNN can realize accurate differential diagnosis between ameloblastoma and odontogenic keratocyst with panoramic radiographs,with higher diagnostic accuracy than oral radiologists.

关 键 词：卷积神经网络 深度学习 曲面体层片 成釉细胞瘤 牙源性角化囊肿 

分 类 号：R739.82[医药卫生—肿瘤]